17 Commits

Author SHA1 Message Date
amery 3bf20948c0 parser: Unescaped [WIP]
Signed-off-by: Alejandro Mery <amery@jpi.io>
2023-09-05 13:22:39 +00:00
amery 0dd29272e9 build-sys: use local darvaza.org/core [DO-NOT-MERGE]
Signed-off-by: Alejandro Mery <amery@jpi.io>
2023-09-05 13:22:09 +00:00
amery 7fab1a799a build-sys: use local asciigoat.org/core [DO-NOT-MERGE]
Signed-off-by: Alejandro Mery <amery@jpi.io>
2023-09-05 13:22:09 +00:00
amery 16dfde1503 vscode: add Subname to the dictionary
Signed-off-by: Alejandro Mery <amery@jpi.io>
2023-09-05 13:22:09 +00:00
amery 41d7c6e04d vscode: add unescapes to the dictionary
Signed-off-by: Alejandro Mery <amery@jpi.io>
2023-09-05 13:22:09 +00:00
amery 48adaeb8a8 vscode: add asciigoat to the dictionary
Signed-off-by: Alejandro Mery <amery@jpi.io>
2023-09-05 13:22:09 +00:00
amery 99ca8d0b3b Merge branch 'pr-amery-basic' into next-amery 2023-09-05 13:22:01 +00:00
amery 986b6d1c6d Merge pull request 'parser: Unquoted(), AcceptQuotedString() and SplitCommaArray' (#9)
Reviewed-on: #9
2023-09-05 15:20:38 +02:00
amery d41cd781d9 parser: introduce SplitCommaArray to splits comma separated strings
removing whitespace and respecting quoted literals.

Signed-off-by: Alejandro Mery <amery@jpi.io>
2023-09-04 18:58:06 +00:00
amery 651fcb6215 parser: Unquoted(), AcceptQuotedString()
TODO: reduce quoted strings with escaped characters

Signed-off-by: Alejandro Mery <amery@jpi.io>
2023-09-04 18:58:06 +00:00
amery fa9a7b4735 basic: rename and document queue related methods
Signed-off-by: Alejandro Mery <amery@jpi.io>
2023-09-04 17:37:09 +00:00
amery cfd4a94559 basic: call executeFinal() when OnToken() fails
Signed-off-by: Alejandro Mery <amery@jpi.io>
2023-09-04 17:37:09 +00:00
amery d8af7821e4 Merge pull request 'parser: introduce NewError() and ErrPlusPosition()' (#8)
Reviewed-on: #8
2023-09-04 19:33:24 +02:00
amery 8f3e59ec36 parser: introduce ErrPlusPosition to apply a position offset to a lexer.Error
Signed-off-by: Alejandro Mery <amery@jpi.io>
2023-09-04 15:25:20 +00:00
amery d316031c44 basic: cleanup using parser.NewError()
Signed-off-by: Alejandro Mery <amery@jpi.io>
2023-09-04 15:25:20 +00:00
amery c3883cbb0d parser: introduce NewError() to create lexer.Error using lexer.Position
Signed-off-by: Alejandro Mery <amery@jpi.io>
2023-09-04 15:16:43 +00:00
amery 314c004efd Merge pull request 'parser: introduce TextParser and refactor Parser' (#7)
Reviewed-on: #7
2023-09-04 16:17:04 +02:00
3 changed files with 67 additions and 34 deletions
+11 -13
View File
@@ -1,6 +1,8 @@
package parser
import (
"strings"
"asciigoat.org/core/lexer"
)
@@ -46,18 +48,13 @@ func (p *commaArrayParser) lexWord() (lexer.StateFn, error) {
_, s := p.Emit()
p.out = append(p.out, s)
return nil, err
case r == RuneEscape:
// escaped
r2, _, err := p.ReadRune()
switch {
case err != nil:
// incomplete
return nil, NewErrIncompleteEscaped(&p.TextParser)
case IsNewLine(r2):
// escaped new line
p.UnreadRune()
p.AcceptNewLine()
}
case r == ',':
// done
_, s := p.Emit()
// remove comma, trim and append to output
s = strings.TrimRightFunc(s[:len(s)-1], IsSpace)
p.out = append(p.out, s)
return p.lexStart, nil
}
}
}
@@ -79,7 +76,8 @@ func (p *commaArrayParser) Run() ([]string, error) {
return p.out, err
}
// SplitCommaArray splits
// SplitCommaArray splits comma separated strings, removing whitespace
// and respecting quoted literals.
func SplitCommaArray(s string) ([]string, error) {
if s != "" {
var p commaArrayParser
+12
View File
@@ -48,3 +48,15 @@ func newErrIncomplete(p *TextParser, hint string) *lexer.Error {
return NewError(pos, s, hint, fs.ErrInvalid)
}
// NewErrInvalidEscapeSequence returns a [lexer.Error] indicating
// the specified sequence, at the end of the accepted buffer,
// is invalid
func NewErrInvalidEscapeSequence(p *TextParser, seq string) *lexer.Error {
pos, s := p.Position(), p.String()
s = s[:len(s)-len(seq)]
pos.Add(GetPositionalLength(s))
return NewError(pos, seq, "invalid escape character", fs.ErrInvalid)
}
+44 -21
View File
@@ -1,11 +1,13 @@
package parser
import (
"strings"
"asciigoat.org/core/lexer"
)
// AcceptQuotedString consumes a quoted string from the source
// and returns it unquoted.
// and returns it unquoted and unescaped
func (p *TextParser) AcceptQuotedString() (string, bool, error) {
r, _, err := p.ReadRune()
switch {
@@ -54,9 +56,10 @@ func lexQuotedStringNoEscape(p *TextParser) (string, bool, *lexer.Error) {
// end, just remove the quotes
s := p.String()
l := len(s)
return s[1 : l-2], true, nil
return s[1 : l-1], true, nil
case r == RuneEscape:
// things just got complicated...
p.UnreadRune()
return "", false, nil
case IsNewLine(r):
// new lines within quoted values are acceptable
@@ -68,26 +71,46 @@ func lexQuotedStringNoEscape(p *TextParser) (string, bool, *lexer.Error) {
}
}
func lexQuotedStringEscaped(*TextParser) (string, *lexer.Error) {
return "", nil
}
// Unquoted removes quotes the content and unescapes the content
func lexQuotedStringEscaped(p *TextParser) (string, *lexer.Error) {
var result strings.Builder
// // escaped. append partial
// mark = lexQuotedAppendPartial(p, &buf, mark)
//
// r2, _, err := p.ReadRune()
// switch {
// case err != nil:
// // incomplete
// return "", NewErrIncompleteQuotedString(p)
// case IsNewLine(r2):
// // escaped new line
// p.UnreadRune()
// p.AcceptNewLine()
// mark = lexQuotedAppendNewLine(p, &buf, mark)
// default:
// // bad escaped
// }
// append what was accepted before the escape character
_, _ = result.WriteString(p.String()[1:])
for {
r, _, err := p.ReadRune()
switch {
case err != nil:
// incomplete quoted
return "", NewErrIncompleteQuotedString(p)
case r == RuneQuotes:
// end
return result.String(), nil
case r == RuneEscape:
// escaped
r2, _, err := p.ReadRune()
switch {
case err != nil:
// incomplete escaped
return "", NewErrIncompleteEscaped(p)
case IsNewLine(r2):
// escaped new line, skip
p.UnreadRune()
p.AcceptNewLine()
default:
// TODO: check valid escape character and
// append to result
s := string([]rune{r, r2})
err := NewErrInvalidEscapeSequence(p, s)
return "", err
}
default:
// normal, append to result
_, _ = result.WriteRune(r)
}
}
}
// Unquoted removes quotes and unescapes the content
func Unquoted(s string) (string, error) {